with the expectation of impersonalism in modern bureaucracies.
The problem of personalism, as Kevin O’Brien and Lianjiang Li have made clear, was exacerbated by the adoption of the “one-level down” management system in 1984. This new system encouraged cadres to be “hyper-responsive to their immediate supervisors at the expense of other interests.” At the same time, the end of widespread political campaigns has weakened party disciple and enhanced cadre autonomy.2 Efforts to implement a civil service system have been largely unsuccessful, largely because civil servants are subject to the cadre management system.3
The cadre management system, based on hierarchical control, has real consequences for how the system works. As Zhou Xueguang has argued, one consequence is that the hierarchical nature of the system compels lower level cadres to engage in collusive behavior. As he writes, “facing unrealistic policy targets and strong incentive pressures, local officials develop coping strategies in the form of collusive behavior to ‘manufacture’ records and statistics to meet policy targets or to transfer resources from one policy area to accomplish the impending tasks in another area.”4 As administrative chains lengthen, the slippage between the demands of higher ups and the performance of local officials inevitably increases.
Another consequence of the cadre management system is that cadres trying to meet the demands of higher ups inevitably confront local residents who have different expectations. Throughout most of the reform era, local cadres have been under pressure to speed up economic growth. In the 1980s and 1990s, local cadres tried to reconcile the demand for higher growth with their shortfall in economic resources by overtaxing the peasants. Whereas central policy limited taxes to 5% of peasant income, local cadres often collected 30–40%. The inevitable result was “rightful resistance.”5 Since the tax reform in 1994 further restricted local resources and especially after the agricultural tax and miscellaneous fees were eliminated in 2006, local officials turned to the requisitioning of land to attract investment. Again, the result was “mass incidents,” which increased year by year. This is to say, the increase of mass incidents is a direct result of the cadre responsibility system.6
A third consequence is corruption and selective policy implementation. A basic goal of any organization is to bring organizational goals and performance incentives into alignment. Throughout much of the period of reform and opening this has meant higher levels pushing for economic development and allowing local officials a great deal of latitude in how to do it. A certain (but unknowable) amount of graft has been part and parcel of this system. Certainly, when Li Changping went to Qipan county he quickly found out that he was expected to participate in corruption; indeed, his refusal to accept bribes was a reason for cadres not to trust him.7 Perhaps, the worst sort of corruption, from an institutional standpoint, is the buying and selling of office. The data for how much of buying and selling of office occur in China is not clear, but certainly it is a non-trivial amount. Jiangnan Zhu’s careful look at the case of Heilongjiang, in which more than 100 officials were involved in the buying and selling of office, argues that because party secretaries have the decisive say on promotions, they have both the ability and incentive to profit from the sale of offices. In contrast to those who argue that China’s political system is undergoing institutionalization, Zhu notes that the buying and selling of office “de-institutionalizes” the state.8
Aligning incentives is not always easy. When the central state is focused on economic growth, this goal aligns very nicely with the interests of local cadres, whether in terms of their professional development (chances for promotion) or personal interests (corruption). It is difficult, in the context of the cadre management system, to “fine-tune” or multiply policy goals. One of the reasons why the local cadres engage in collusive behavior, Zhou Xueguang points out, is that the criteria on which their behavior is being judged is not always clear. If higher-ups have multiple goals — say, environmental protection, economic growth and the “well-being” of the population — understanding the criteria for promotion becomes even more difficult, and the personalization of bureaucratic relations intensifies. One way to deal with this uncertainty over goals is for the local cadres to pick the policy goals that he believes the superordinate cares about most — especially if that goal aligns with his own personal interests. Thus, there is a systemic tendency toward selective policy implementation.9
The problem of alignment between the goals of the center and the interests of the local cadres becomes particularly severe when they diverge. The policy of fangquangrangli 放权让利 that was followed in the 1980s was very good at stimulating economic growth, but it also setup centripetal forces that loosened the center’s control. The tax-sharing system that was implemented in 1994 and the elimination of the agricultural tax and miscellaneous fees in 2006 were intended to strengthen the center’s control (which they did in fiscal terms), but they deprived the localities of revenues while still expecting them to develop the economy. It is not surprising that local cadres found new sources of revenue, exacerbating cadre-mass relations and worsening the problem of corruption.
In recent years, there has been a debate about whether promotions are based on personalistic considerations (such as patron–client ties) or on merit. Given that local level cadres find the criteria for promotion less than clear, that the differentiation between levels of the bureaucracy is not large enough to constrain behavior, and that misbehavior is rarely sanctioned, the incentives to develop informal ties and to promote people on the basis of them, at least in part, are very strong. The evidence of the sale of office also suggests that control over promotion is highly concentrated and not subject to effective oversight. Although merit may enter consideration for promotion, it seems pretty clear that it is not the only consideration.
I think that one way to conceive of the difference between the cadre system in China and a Weberian-style bureaucracy is to think of the difference between “cadres” and “administrators.” Is the term “cadre” simply a translation of the term “administrator” or “bureaucrat”? I don’t think that anyone who has looked at the Chinese political system would argue that it is. What distinguishes a cadre from a bureaucrat is first of all the very personal relationship between the superordinate and the subordinate. Ultimately, it is the superordinate who judges the behavior of the subordinate, and this makes satisfying the demands of the superordinate the primary consideration, guiding the subordinate’s behavior. Second, and in tension with this first characteristic, the subordinate is given a great deal of leeway to achieve whatever tasks have been assigned. Note the difference between “tasks” and “job descriptions”; bureaucrats have job descriptions, cadres have tasks. Tasks are policy goals, such as “grow the economy” and do not necessarily come with the resources necessary to accomplish them. It is up to the subordinate to figure out how to complete the tasks. And that means that the subordinate must have a great deal of discretion, indeed, of arbitrary power. Without that arbitrary power, the cadre would just be a bureaucrat. This suggests that in contrast to the separation of person and office that characterizes Weberian-style bureaucracies, that person and office really are one and the same, at least during the period in which one holds office. This is really quite different either from traditional systems in which one holds office by virtue of characteristics possessed by the individual (such as aristocratic blood) and from Weberian-style separation of person and office. As long as the cadre system holds, cadres can perform their jobs more or less well and more or less honestly, but they will not become bureaucrats. And the party system will not become a Weberian-style bureaucracy.
All this raises an interesting question about the changes that seem to be going on now. There are two changes that seem to affect the party system. The first is the separation of the local court system out from the party committee at the same level, with finances coming directly down from the province. The idea would be that this would make court judgments less susceptible to interference and hence more predictable. The other is that the Discipline Inspection Commissions one level above are to take charge of corruption investigations, thus tightening vertical supervision. These reforms are intended to reduce the scope of the local party secretary’s power. Will this turn cadres into bureaucrats? I don’t
